FinAI / SecureFinAI Lab (Columbia) with the Linux Foundation, FINOS and Hugging Face, 2025
Open FinLLM Leaderboard
A public leaderboard of LLMs on financial tasks. The original Space covered seven FinBen-style categories. The current TheFinAI board shows MultiFinBen results across five languages and text, vision and audio.
Best reported
46.0
GPT-4oOpenAI
balanced score (mean of text/vision/audio averages)
Official board, 16 Apr 2026
About this benchmark
- Size
- Current board: 21 models on 36 datasets. The paper says 42 datasets in 7 categories.
- Public split
- Scores (scores.json) and the Space source are public. Datasets are on Hugging Face under their own cards.
- Grading
- Zero-shot automatic metrics per dataset (accuracy, F1, ROUGE, MCC). The current board ranks by a balanced score: the mean of the text, vision and audio averages.
- Licence
- Apache-2.0 (Space code); the paper says the project follows the OpenMDW License
- Submission
- via-maker
- Products allowed
- No
- Latest version
- Board data updated 2026-04-16 (MultiFinBen arXiv 2506.14028v3, Table 3)
- What it does not cover
- No agents or tools, no live retrieval, and few frontier closed models (GPT-4o is the only closed model in the top 10).
Results
10 modelsEach score shows who reported it: the benchmark's authors (quoted from their paper), the benchmark's official board, or Moneda (run with the benchmark's own grader and rules). We do not copy other aggregators.
Open FinLLM Leaderboard Board data updated 2026-04-16 (MultiFinBen arXiv 2506.14028v3, Table 3)
balanced score (mean of text/vision/audio averages), higher is better. 10 models.
Updated 16 Apr 2026
- GPT-4o46.0
- Qwen2.5-Omni-7B33.5
- Llama-4-Scout-17B-16E20.9
- Whisper-V317.2
- Qwen2-Audio-7B-Instruct16.7
- Qwen2-Audio-7B16.0
- Gemma-3-27B15.7
- Llama-3.1-70B14.1
- Gemma-3-4B12.7
- Qwen2.5-32B12.7
Open FinLLM Leaderboard Board data updated 2026-04-16 (MultiFinBen arXiv 2506.14028v3, Table 3)
balanced score (mean of text/vision/audio averages), higher is better. 10 models.
Updated 16 Apr 2026
- 1GPT-4o46.0
- 2Qwen2.5-Omni-7B33.5
- 3Llama-4-Scout-17B-16E20.9
- 4Whisper-V317.2
- 5Qwen2-Audio-7B-Instruct16.7
- 6Qwen2-Audio-7B16.0
- 7Gemma-3-27B15.7
- 8Llama-3.1-70B14.1
- 9Gemma-3-4B12.7
- 10Qwen2.5-32B12.7
| 1 | GPT-4o | OpenAI | 46.0 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 |
|---|---|---|---|---|---|
| 2 | Qwen2.5-Omni-7B | Qwen | 33.5 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 |
| 3 | Llama-4-Scout-17B-16E | Meta | 20.9 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 |
| 4 | Whisper-V3 | OpenAI | 17.2 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 |
| 5 | Qwen2-Audio-7B-Instruct | Qwen | 16.7 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 |
| 6 | Qwen2-Audio-7B | Qwen | 16.0 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 |
| 7 | Gemma-3-27B | 15.7 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 | |
| 8 | Llama-3.1-70B | Meta | 14.1 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 |
| 9 | Gemma-3-4B | 12.7 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 | |
| 10 | Qwen2.5-32B | Qwen | 12.7 | MultiFinBen Table 3 data | Open FinLLM Leaderboard (scores.json) Official board, reported 16 Apr 2026 · checked 1 Nov 2026 |
Papers and links
Last checked 1 Nov 2026. Something out of date or wrong? Send a correction.